The Canadian Field-Naturalist

Banded ( diaphanus) and (Fundulus heteroclitus) distributions in insular Newfoundland waters: implications for a at Risk

Philip S. Sargent1, *, Kate L. Dalley1, and Derek R. Osborne1

1Department of Fisheries and Oceans Canada, Northwest Atlantic Fisheries Centre, 80 East White Hills Road, St. John’s, Newfoundland and Labrador A1C 5X1 Canada *Corresponding author: [email protected]

Sargent, P.S., K.L. Dalley, and D.R. Osborne. 2020. (Fundus diaphanus) and Mummichog (Fundus hetero­ clitus) distributions in insular Newfoundland waters: implications for a Species at Risk. Canadian Field-Naturalist 134(4): 307–315. https://doi.org/10.22621/cfn.v134i4.2373

Abstract Newfoundland’s Banded Killifish Fundulus( diaphanus) population is listed as a species of Special Concern under Canada’s Species at Risk Act and Vulnerable under Newfoundland and Labrador’s Endangered Species Act. Mummichog (Fundulus heteroclitus) is a similar looking species and is currently under review by Newfoundland and Labrador’s Species Status Advisory Committee. Both species have limited known distributions in Newfoundland waters that overlap. They may occur sympatrically in and occasionally hybridize; thus, field identifications can be challenging. We found that dorsal fin position and caudal fin depth were the most useful morphological characters for distinguishing Banded Killifish and Mummichog in the field. We used local ecological knowledge, literature review, museum records, and field surveys to update the known distribution ranges and found both species in more locations than previously documented in Newfoundland. Thus, we extend their known ranges. Our results will be critical in future status assessments of these species in Newfoundland. Key words: Banded Killifish; distribution range; Fundulus; identification; Mummichog; Newfoundland; range extension; species at risk Introduction the Newfoundland population of Banded Killifish Oviparous cyprinodontiform , commonly were assessed as species of Special Concern by the known as and topminnows, occur - natu Committee on the Status of Endangered Wildlife rally on all continents except Australia and Antarctica in Canada (COSEWIC 2012, 2014) and are listed and are common in fresh, brackish, and occasion- as such under Canada’s Species At Risk Act (SARA ally coastal seawater (Scott and Scott 1988). In North Registry 2019a,b). The Newfoundland population of America, they extend as far north as southern Canada Banded Killifish is also listed as Vulnerable under where there are three Fundulus species (family Fun­ Newfoundland and Labrador’s Endangered Species dulidae):­ Blackstripe Topminnow (Fundulus notatus), Act (Endangered Species List Regulations 2002). Banded Killifish Fundulus ( diaphanus), and Mum­ Throughout most of its distribution range Banded michog (Fundulus heteroclitus; Houston 1990). There Killifish is considered Not at Risk (COSEWIC 2014; are two subspecies of Banded Killifish, the east- SARA Registry 2019b). The Newfoundland popula- ern subspecies (Fundulus diaphanus diaphanus) and tion’s Special Concern and Vulnerable designations the western subspecies (Fundulus diaphanus me­ are due to a limited and clustered distribution in insular nona), both of which are present in Canada (Scott and Newfoundland, which makes them vulnerable to cat- Crossman 1973) but only the eastern subspecies oc- astrophic events and local disturbances (COSEWIC curs in Newfoundland. There are also two subspecies 2014). Also present in Newfoundland waters is Mum­ of Mummichog, the southern subspecies (Fundulus michog (Scott and Crossman 1973), whose status is heteroclitus heteroclitus) and the northern subspecies currently under review by the Newfoundland and (Fundulus heteroclitus macrolepidotus), of which Labra­dor Species Status Advisory Committee (T. only the latter occurs in Canadian waters (Able and Knight pers. comm. 28 January 2016). Literature sug- Felley 1986). In Canada, Blackstripe Topminnow and gests Mummichog distribution is restricted to south-

307 ©Fisheries and Oceans Canada 2021. This work is freely available under the Creative Commons Attribution 4.0 International license (CC BY 4.0). 308 The Canadian Field-Naturalist Vol. 134 western Newfoundland (Scott and Scott 1988). cies (Chippett 2004). Localized surveys for Banded Banded Killifish (Figure 1a) and Mummichog Killifish, conducted just prior to the 2003 COSEWIC (Figure­ 1b) are both considered , but Banded assessment and update status report (in Terra Nova Killifish is more of a freshwater species, whereas National Park [Cote et al. 2002]; Gros Morne Mum­michog is more of a brackish-water species with National Park [Knight 2002]; Indian Bay watershed only a few documented freshwater populations (Klawe [Chippett 2004]), added only one watershed to their 1957; Denoncourt et al. 1978; Scott and Scott 1988). known distribution. Since 2006, the Mi’kmaq Alsumk Both species are very similar in appearance and may Mowimsikik Koqoey Association (MAMKA) doc- school together where sympatric populations occur in umented by-catch of Banded Killifish from the brackish waters (Scott and Crossman 1964). In addi- American Eel (Anguilla rostrata) commercial fish- tion, these two species have been reported to occa- ery in western Newfoundland (MAMKA 2006) sionally hybridize (Fritz and Garside 1974; Dawley and conducted Banded Killifish surveys (MAMKA 1992). These factors make field identification - diffi 2011). However, except for a few earlier reports cult, which may affect population studies and delinea- (e.g., Templeman 1951; Scott and Crossman 1964; tion of their respective distribution ranges (Fisheries Day 1993), there was no direct evidence of Banded and Oceans Canada 2011). Killifish at many of the locations presented in Little effort has been spent in delineating the dis- COSEWIC (2014). tribution range of Banded Killifish in Newfoundland. The objectives of this study were to determine the Most early reports have been accidental discover­ morphometric characters that allow for clear differ- ies by anglers and researchers studying other spe­ entiation of Banded Killifish and Mummichog in the field and to update their known distribution ranges in insular Newfoundland waters. Field surveys, litera- ture reviews, museum records, and local ecological knowledge (LEK) were used to update distribution ranges. Results from this study will provide new data that will help in the assessment on their listing status both at the federal and provincial levels. Methods A literature search was conducted for records of Banded Killifish and Mummichog in Newfoundland. Museums, including the Canadian Museum of Na­ ture (CMN, Ottawa, Ontario [ON]), Royal Ontario Mu­seum (ROM, Toronto, ON), Atlantic Reference Centre (ARC, St. Andrew’s, New Brunswick [NB]), The Rooms Natural History Department (The Rooms, St. John’s, Newfoundland and Labrador [NL]), Mau­ rice Lamontagne Institute (MLI, Mont-Joli, Quebec [QC]), Ministère des forêts, de la faune et des parcs du Québec (Longueuil, QC), and Nova Scotia Mu­ seum (Halifax, Nova Scotia), were also contacted for reports of Banded Killifish. Federal fisheries -of ficers from Fisheries and Oceans Canada, person- nel from the Provincial Departments of Fisheries and Aquaculture, and Environment and Conservation, MAMKA river guardians, commercial eel harvesters, academic researchers, and local residents were con- tacted via telephone, e-mail, or in-person to gather lo- cal knowledge and determine potential locations of both species in Newfoundland. Pictures of the spe- cies were provided, and people contacted were asked if and where they had been observed. Maps were pro- Figure 1. a. (i) Female and (ii) male Banded Killifish Fun( ­ dulus diaphanus); b. (i, iv, v) male and (ii, iii) female Mum­ vided to assist the identification of drainage systems michog (Fundulus heteroclitus), each exhibiting variations where Fundulus spp. were observed, where a ‘drain- in banding patterns. Photos: Kate Dalley. age system’ was defined as any water system with 2020 Sargent et al.: Newfoundland killifish distributions 309 a separate drainage to the ocean. Several commer- the ARC (St. Andrew’s, NB) for identification were cial eel harvesters were requested to retain Banded deposited there (Table S1). Sampling data from this Killifish/Mummichog specimens captured in their study has been submitted to the Ocean Biogeographic fishing gear and, in some instances, incidental conver- Information System (OBIS) for public archive. Reports sations with locals during field surveys were used to of each species were verified using physical specimens gather additional information. Information gathered or detailed photographs (Figure S1). from the aforementioned sources was used to plan In the laboratory, meristic characters (Scott and field surveys and focus effort in general areas where Crossman 1964) and one morphological character there were reports of these species. (Scott and Scott 1988) were used to verify field iden- A total of 102 sites were sampled for Banded tifications. Meristic characters (Table 1) included the Killifish and Mummichog from 2013 to 2018. Most number of dorsal fin rays, the number of gill rakers were sampled using four Gee Minnow Traps (42 cm L on the first gill arch, and the number of scale rows, × 19 cm D, 22 mm opening, 6.4 mm mesh; Fillmore, and the stepped forward location (SFL) morphometric New York, USA) that had been presoaked in saltwa- were used to identify the species (Table 1; Figure 2). ter for 24 h to remove the surface shine. Traps were Smaller juveniles were difficult to identify, and several each baited with 10 Original Ritz crackers (Mondelēz other individuals showed a mix of characteristics sug- International, Toronto, Ontario, Canada). Additional gesting the possibility of hybridization and were sent opportunistic collection methods included minnow to L. van Guelpen (ARC) for species identification. traps baited with sardines, fyke nets, hoop nets, dip After initial identification of specimens collected nets, pole seines, LR-24 Electrofisher (Smith-Root, from 2013 to 2016, additional morphological mea- Vancouver, Washington, USA), and from stomach surements were recorded as potential characters to contents of Brook (Salvelinus fontinalis) caught differentiate species for individuals ≥27.5 mm SL in icefishing with baited hook and line, while other spec- the field (Figure 2). Morphological characters identi- imens were provided by local residents. Catches from fied from the literature to differentiate these species -in each survey were identified, counted, and standard cluded the dorsal fin index (DFI; Scott and Crossman length (SL) measured. Samples of Fundulus spp. (typ- 1964; Table 1; Figure 2) and the ratio of the caudal ically 4–6 individuals) were preserved in 70% ethanol peduncle depth (CD) relative to the distance from from each location for later analysis of morpholog- the dorsal origin to posterior end of vertebrae (DO– ical and meristic characters for species identifica- EV); the inverse of the ratio used by Fritz and Garside tion and museum archival. Most Banded Killifishand (1974) and Hernández Chávez and Turgeon (2007). Mummichog specimens retained in this study were de- Data for Banded Killifish, Mummichog, and pos- posited at The Rooms, Provincial Museum of Natural sible hybrids were plotted using SigmaPlot version History Annex (St. John’s, NL), whereas those sent to 13.0 (Systat 2014) to determine the amount of over-

Table 1. Definitions of meristic and morphometric characters used to identify Banded Killifish (Fundulus diaphanus) and Mummichog (Fundulus heteroclitus) in Newfoundland. See Figure 2 for illustrations of the morphometric measurements.

Character Character Description Meristic Number of dorsal fin rays (DFR) Number of fin rays on the dorsal fin Meristic Number of gill rakers (GR) Number of gill rakers on the first gill arch Meristic Number of scale rows (SR) Number of scales in the longitudinal row from just posterior to the oper- culum to the end of the caudal peduncle Morphometric Caudal depth (CD) Vertical distance from the dorsal to the ventral part of the caudal peduncle Morphometric Dorsal origin to posterior end Distance from the anterior origin of the dorsal fin to the posterior end of vertebrae (DO–EV) of the vertebral column Morphometric Dorsal origin to tip of snout Distance from the anterior origin of the dorsal fin to the tip of the snout (DO–ST) Morphometric Standard length (SL) Distance from the tip of the snout to the posterior end of the vertebral column Morphometric Dorsal fin index (DFI) Measure the DO–EV distance and subtract it from the DO-ST distance (see Scott and Crossman 1964) Morphometric Stepped forward location (SFL) Measure the DO–EV distance and step that distance anteriorly from the anterior origin of the dorsal fin; the location on the head at which this measurement lands determines the species; if the location lands near the eye it was identified as Banded Killifish, whereas if it landed on the operculum it was identified as Mummichog (Scott and Scott 1988: 612) 310 The Canadian Field-Naturalist Vol. 134

Two adult specimens from Little Paradise Park (>50 mm SL) and eight juvenile specimens (<20 mm SL) from Saltwater Pond sent to the ARC for identifi- cation could only be identified asFundulus sp. (Table S2; Figure S2). The juveniles from Saltwater appeared to be Mummichog but the adults from Little Paradise Park exhibited a mix of characters from both Figure 2. Morphological measurements recorded from kil- species (L. van Guelpen pers. comm. 13 June 2014; lifish (Fundulus spp.) in insular Newfoundland. Abbre­vi­ Figure S2). ations: CD = caudal depth; DO–EV = origin of dorsal fin to We confirmed the presence of Banded Killifish at end of vertebrae; DO–ST = origin of dorsal fin to snout tip; SL = standard length. 45 sites within 35 drainage systems (Figure 3) and Mummichog at 30 sites within 24 drainage systems (Figure 4) in insular Newfoundland (Tables 2 and lap of each character. Characteristics with <5% over- S1). From the 102 sites surveyed during this study, lap were identified as the most useful for differentiat- 30 and 18 were new (i.e., previously undocumented ing Banded Killifish and Mummichog. or unconfirmed from LEK and grey literature, such Results as internal reports; Table S1) for Banded Killifish and In general, Banded Killifish (Figure 1a) has a more Mummichog, respectively. Locations were considered slender and compressed body form with a more ta- unconfirmed when grey literature and LEK lacked pered mouth compared to Mummichog (Figure 1b). sufficient physical evidence for accurate species iden- Mummichog has a stout, robust body form and deeper tification. Banded Killifish and Mummichog were not caudal peduncle compared to Banded Killifish. Fe­ detected at 70 and 82 sites, respectively (Figures S3 males of both species exhibit thin vertical black bands and S4). Potential locations of Banded Killifish and along their sides, but on Mummichog, these bands, Mummichog occur where unconfirmed reports were posterior to the anal fin, are usually shortened and do not investigated (Figures S3 and S4). Banded Killifish not span the width of the body (Figure 1). Males of and Mummichog were detected sympatrically at two both species do not have black bands but instead have unnamed that connect directly to estuaries at alternating dark olive and lighter white/silver/blue high : one near Little Paradise Park, St. Andrew’s bands during the breeding season that are closer to- and the other near Stephenville Crossing (Table S1). gether compared to bands of females (Figure 1). When present, catch numbers from minnow traps From our measurements and analyses (Table S2; ranged up to 102 individuals with catch per unit effort Figure S2), we developed the following identification (CPUE) from 0.04 to 16.92 fish/h for Banded Killifish key from the most useful meristic and morphomet- and up to 159 individuals with CPUE from 0.04 to ric characters. Individuals that exhibited characters 12.52 fish/h for Mummichog. that when keyed out did not clearly identify as either We found two unreported museum records for Banded Killifish or Mummichog were considered po- Banded Killifish (Burin and Gravels Pond; Table tential hybrids. S1) and one unreported record for Mummichog Number of gill rakers on the first gill arch, 4–7 (usu- (Terrenceville; Table S1). We also detected two er- ally five); number of scale rows, 42–55; stepped roneous reports of Banded Killifish from Star forward location reaches just anterior to the eye and York Harbour, Newfoundland. The Star Lake re- to between the eye and operculum (usually mid- cord was reported by Chippett (2004) based on a per- dle of the eye); ratio of caudal peduncle depth rel- sonal communication, but we found that the origi- ative to the distance between the origin of the dor- nal source indicated Banded Killifish was actually sal fin and the end of the vertebrae, 0.15–0.25 ...... observed in Stag Lake (B. Dennis pers. comm. 9 ...... Banded Killifish,Fundulus diaphanus January 2015). We sampled Stag Lake but could not Number of gill rakers on the first gill arch 6–10 (usu- confirm the presence of Banded Killifish. Fisheries ally eight); number of scale rows, 32–39; stepped and Oceans Canada (2011) and COSEWIC (2014) forward location reaches posterior to the eye to reported Banded Killifish from York Harbour, how- posterior to the operculum (usually posterior oper- ever, no specimens were retained from the original culum); ratio of caudal depth relative to the dis- sampling, only photographs. Morphological features tance between the origin of the dorsal fin and the measured from these photos (Figure S1d–f) indicated end of the vertebrae, 0.25–0.40 ...... Mummichog, not Banded Killifish. In addition, our ...... Mummichog, Fundulus heteroclitus surveys at this site yielded only Mummichog. 2020 Sargent et al.: Newfoundland killifish distributions 311

Figure 3. Confirmed locations of Banded KillifishFundulus ( diaphanus) in insular Newfoundland.  = locations confirmed through direct sample collection or samples provided by residents during this study.  = locations confirmed by museum and literature records, or unpublished data with substantial evidence of species identification (e.g., high quality photographs).

Figure 4. Confirmed locations of Mummichog (Fundulus heteroclitus) in insular Newfoundland.  = locations confirmed through direct sample collection or samples provided by residents during this study.  = locations confirmed by museum and literature records, or unpublished data with substantial evidence of species identification (e.g., high quality photographs). 312 The Canadian Field-Naturalist Vol. 134

Table 2. Sources of site confirmations of Banded KillifishFundulus ( diaphanus) and Mummichog (Fundulus heteroclitus) in insular Newfoundland.

Literature* with Surveys Museum LEK Surveys Specimens Literature* corresponding confirm- Surveys Species records with confirming provided Totals only museum ing grey alone only photos LEK by locals records literature Banded Killifish 7 4 2† 2‡ 10 0 2 18 45 Mummichog§ 3 5 3 0 7 6 1 5 30 *Literature sources included: Johansen (1926); Templeman (1951); Scott and Crossman (1964); van Vliet (1970); Dickinson and Threlfall (1975); Gibson et al. (1984); Day (1993); Chippett (2004); Mann and Nambudiri (2005); Mitchell and Purchase (2014). †One site (Burin) could not be counted in total as site co-ordinates were not provided with sample. ‡See Figure S1a–c used to confirm Banded Killifish reports from LEK (local ecological knowledge). §Site confirmations based on LEK and grey literature (internal documents) were for reports believed to be Banded Killifish.

Discussion COSEWIC (2014). Two new locations were consider- As previously identified by Scott and Crossman ably further north (77 and 146 km respectively), than (1964) and Hernández Chávez and Turgeon (2007), previously reported (COSEWIC 2014) and we doc- the number of scale rows along the was umented them in one additional drainage system on the most useful meristic character to differentiate the northeast where only one was previously re- these species, because there is no overlap. We found ported (Chippett 2004). We also corrected two Banded the stepped forward position and the ratio of caudal Killifish locations previously misreported by Chippett depth to the distance between the origin of the dorsal (2004) and Fisheries and Oceans Canada (2011) due fin and the end of the vertebrae to be the most effec- to a miscommunication of a location and a misiden- tive morphometric characters for differentiating these tification of Mummichog, respectively. Mummichog species in the field. These characters, previously iden- was previously reported at ten locations from eight tified by Scott and Scott (1988) and Fritz and Garside drainage systems along the southwest coast of insular (1974), respectively, only require three measurements Newfoundland (Johansen 1926; Scott and Crossman in the field, minimizing handling stress and even al- 1964; van Vliet 1970; Dickinson and Threlfall 1975; low identifications from high quality lateral view Scott and Scott 1988). We confirmed them at three photographs. times as many sites and drainage systems, including Several of our specimens could only be identified eight sites along the northeast coast where they have as Fundulus sp.; two individuals from Little Paradise never been previously reported. The previously re- Park (Little Codroy River ) were considered ported limited distributions of both Banded Killifish potential hybrids, as the species were sympatric in a and Mummichog was most likely due to a lack of sur- pond that connects to the estuary near the collection vey data; the new locations we identified highlight a site. Hybridization has been documented at two lo- general lack of basic aquatic biodiversity data for in- cations in Nova Scotia (Fritz and Garside 1974) but sular Newfoundland. is probably more widespread (Hernández Chávez and Banded Killifish can colonize new territory and ex- Turgeon 2007). However, hybridization has yet to be pand their range when conditions are suitable. A pop- confirmed in Newfoundland. ulation of Banded Killifish was reportedly introduced Several previously reported locations of Banded into a pond in the city of St. John’s (east coast) in 1999 Killifish are likely in error due to misidentification. (Mitchell and Purchase 2014). In 2014, we detected a Similar to the erroneous York Harbour record, speci- downstream expansion of this population into a lake mens were not retained from locations in West Bay of but did not detect an expansion upstream likely be- the Port au Port Peninsula (MAMKA 2011). We sur- cause of a ~4 m high waterfall, which may present a veyed most of these West Bay sites and several were significant barrier (Gibson et al. 1984). The expan- surveyed earlier by Johansen (1926) and van Vliet sion of this population over the last 20 years suggests (1970) but only Mummichog were detected. resilience and the potential for population restocking Our results greatly expanded the known number (Mitchell and Purchase 2014) elsewhere if deemed of locations for Banded Killifish and Mummichog necessary. Expansion of at-risk populations of Banded in insular Newfoundland and extended their known Killifish in the (Illinois) have also been ranges. We confirmed Banded Killifish at nearly reported (Mankowski 2012). Populations of the eastern four times as many sites in more than three times and the western subspecies have been rapidly expand- as many drainage systems as previously reported in ing in Lake Michigan since 2001 and the Mississippi 2020 Sargent et al.: Newfoundland killifish distributions 313

River since 2009 (Willink et al. 2018). Recent changes imens would have to be examined, as suggested by in environmental conditions within the Great Hernández Chávez and Turgeon (2007). due to and climate change may have benefited the eastern subspecies in this area, while the Author Contributions western subspecies may have been introduced into the Writing – Original Draft: P.S.S.; Writing – Re­ Mississippi River (Willink et al. 2018). view & Editing: D.R.O., K.L.D., and P.S.S.; Concep­ ­ In light of the expanded distribution we found, the tualization: P.S.S.; Investigation: K.L.D. and P.S.S.; status of Newfoundland’s Banded Killifish popula- Methodology: K.L.D. and P.S.S.; Formal Analysis: tion should be reassessed by COSEWIC. This pop- P.S.S.; Data Curation: P.S.S.; Project Administration: ulation was initially designated as Vulnerable (re- P.S.S.; Resources: P.S.S. and D.R.O; Supervision: named Special Concern since 2000; COSEWIC 2003, P.S.S.; Validation: P.S.S. and K.L.D.; Funding Acqui­ 2014) in 1989 (Houston 1989, 1990), based on only sition: P.S.S. two known widely separated localities reported by Acknowledgements Scott and Crossman (1964) and Gibson et al. (1984). Thanks to Gerry Yetman (Wildlife Division, The Special Concern status was maintained in the Newfoundland and Labrador Government [retired]); 2003 and 2014 assessments despite the addition of Melissa Graham (Clarke University); Eric Schultz four locations (COSEWIC 2003) and five locations (University of Connecticut); Paul Wood (Ducks (COSEWIC 2014), two of which (Star Lake and York Unlimited); Ed Webb (Mi’kmaq Alsumk Mowimsikik Harbour) we have found to be in error. Koqoey Association); Nicole Glover (Qalipu First To date, targetted surveys for Banded Killifish have Nation); and Robert Diamond and Belinda House been extremely limited. Including the present study, (commercial eel harvesters) for potential sampling fewer than 200 (<1%) of Newfoundland’s ponds, sites. Special thanks to Ed Webb and Dave Lucas lakes, brooks, and barachois have been surveyed (Mi’kmaq Alsumk Mowimsikik Koqoey Association) specifically for Banded Killifish (see Gibson et al. for advice on Banded Killifish trapping. We thank Janet 1984; Cote et al. 2002; Knight 2002; Chippett 2004; Sargent and Viviana Ramírez Luna (P.S.S. family) for MAMKA 2006, 2011). Given the vast number of wa- field assistance. Sincere thanks to Marty Swyers and ter bodies in insular Newfoundland, future sampling Raymond Young (Mi’kmaq Alsumk Mowimsikik must be prioritized and optimized. Representative Koqoey Association); Melissa Brake, Andrea Coombs, water bodies from each drainage system with suitable and Jonathan Strickland (Qalipu First Nation); environmental conditions for Banded Killifish should Bradley Coles and Craig Hodder (Fishery Guardians); be considered first. We recommend sampling: 1) ar- Raymond Reid, Carson Wentzell, and Darroch Whi­ eas with suitable habitat including shallow quiet wa- taker (Parks Canada); Wally Cunard Jr. (commer- ters of ponds and lakes with a sand, gravel, or detri- cial eel harvester); and Larry Quinlan (local resident) tus-covered bottom and patches of submerged aquatic for providing Fundulus specimens. We thank Bill plants (Scott and Crossman 1973); 2) water bodies Dennis (Department of Fisheries and Aquaculture, in the lowest parts of each drainage system without NL Government); Tom Knight (Parks Canada); and steep gradients that may create barriers to upstream Lou van Guelpen (Atlantic Reference Centre) for migration (Gibson et al. 1984); 3) using minnow traps personal communications. Thanks to Greg Moore baited with Ritz crackers, the most efficient sampling (Atlantic Coastal Action Program-Humber Arm); method for Banded Killifish and Mummichog >30 Emma O’Melia (Clark University); and Eric Schultz mm (SL) during our study; 4) when the water tem- (University of Connecticut) for providing Banded perature is ≥17°C (Chippett 2004), typically between Killifish photographs. We thank the anonymous- re July and mid-September; and 5) during sunny days, viewer and journal editors for refining the manu- when Banded Killifish were more readily observed script. This project was funded by the Species at during our study. Additionally, detection may de- Risk and Small Craft Harbours Programs and Dr. pend on size of the water body surveyed. To increase Robert Gregory (Fisheries and Oceans Canada). Fish chances of detecting Banded Killifish in larger wa- were captured under Fisheries and Oceans Canada ter bodies, sampling effort should be increased pro- Experimental Licences and approved Care portionally to water body size. Future surveys should Protocols. Animal­ care protocols for the handling and also explore the distribution of Banded Killifish on euthanasia of specimens in this study were reviewed the northeast coast, as it is unclear how the species and approved annually by the Animal Care Committee arrived there and how widespread it is in this region. of the Northwest Atlantic Fisheries Centre. To confirm the presence of hybrids in Newfoundland, morphometric and genetic data (both mitochondrial Literature Cited and nuclear sequence polymorphism) from spec- Able, K.W., and J.D. Felley. 1986. Geographical variation 314 The Canadian Field-Naturalist Vol. 134

in Fundulus heteroclitus: tests for concordance between nus in Newfoundland. Le Naturaliste Canadien 111: egg and adult morphologies. American Zoologist 26: 213–214. 145–157. https://doi.org/10.1093/icb/26.1.145 Hernández Chávez, C., and J. Turgeon. 2007. Asexual Chippett, J.D. 2004. An examination of the distribution, hab- and sexual hybrids between Fundulus diaphanus and F. itat and genetic and physical characteristics of Fundulus­ heteroclitus in the Canadian Atlantic region. Molecu­ ­ diaphanus, the Banded Killifish, in foundlandNew­ and lar Ecology 16: 1467–1480. https://doi.org/10.1111/j.13 Labrador.­ M.Sc. thesis, Memorial University of Newfound­­ 65-294X.2007.03239.x land, St. John’s, Newfoundland and Labrador, Canada.­ Houston, J.J.P. 1989. COSEWIC status report on the COSEWIC (Committee on the Status of Endangered Banded Killifish Fundulus diaphanus in Canada. Com­ Wildlife in Canada). 2003. COSEWIC assessment and mittee on the Status of Endangered Wildlife in Canada, update status report on the Banded Killifish Fundulus Ottawa, Ontario, Canada. dia­phanus, Newfoundland population in Canada. COSE­ Houston, J. 1990. Status of the Banded Killifish, Fundulus WIC, Ottawa, Ontario, Canada. diaphanus, in Canada. Canadian Field-Naturalist 104: COSEWIC (Committee on the Status of Endangered 45–52. Accessed 11 March 2014. https://www.biodiversity Wildlife in Canada). 2012. COSEWIC assessment and library.org/page/34346593. update status report on the Blackstripe Topminnow Fun­ Johansen, F. 1926. Fishes collected in Newfoundland during du­lus notatus in Canada. COSEWIC, Ottawa, Ontario,­ the autumn of 1922. Canadian Field-Naturalist 40: 31– Canada. 36. Accessed 15 October 2020. https://www.biodiversity COSEWIC (Committee on the Status of Endangered library.org/page/28025291. Wild­life in Canada). 2014. COSEWIC assessment and Klawe, W.L. 1957. Common Mummichog and newt in a lake status report on the Banded Killifish Fundulus diapha­ on Digby Neck, Nova Scotia. Canadian Field-Naturalist nus in Canada. COSEWIC, Ottawa, Ontario, Canada. 71: 154–155. Accessed 7 January 2015. https://www.bio Cote, D., M. Langdon, R. Collier, G. Sparkes, and D.A. diversitylibrary.org/page/28059827. Scru­ton. 2002. The status of the eastern Banded Killifish Knight, T.W. 2002. The distribution and status of the eastern Fun­dulus diaphanus in Terra Nova National Park. Parks Banded Killifish, Fundulus diaphanus, in Gros Morne Can­ada, Glovertown, Newfoundland and Labrador, Can­ National Park of Canada, Newfoundland. Parks Canada, ada. Rocky Harbour, Newfoundland and Labrador, Canada. Dawley, R.M. 1992. Clonal hybrids of the common labo- MAMKA (Mi’kmaq Alsumk Mowimsikik Koqoey Asso­ ratory fish Fundulus heteroclitus. Proceedings of the ciation). 2006. Banded Killifish: monitoring the by- National Academy of Sciences of the United States of catch in the eel fishery. MAMKA, Corner Brook, New­ America 89: 2485–2488. https://doi.org/10.1073/pnas. foundland and Labrador, Canada. 89.6.2485 MAMKA (Mi’kmaq Alsumk Mowimsikik Koqoey Asso­ Day, R. 1993. Ramea Islands: additions to the flora, fish & ciation). 2011. Banded Killifish dispersal in insular fauna. Osprey 24: 168. New­foundland. Draft Report. MAMKA, Corner Brook, Denoncourt, R.F., J.C. Fisher, and K.M. Rapp. 1978. A Newf­oundland and Labrador, Canada. freshwater population of the Mummichog, Fundulus Mankowski, A. 2012. The Illinois endangered species pro­ hete­roclitus, from the Susquehanna River drainage in tection act at forty: a review of the act’s provisions Pennsylvania. Estuaries 1: 269–272. https://doi.org/10.23 and the Illinois list of endangered and threatened spe- 07/1351534 cies. Endangered Species Protection Board, Springfield, Dickinson, A.B., and W. Threlfall. 1975. Metazoan para- Illinois, USA. sites of Fundulus heteroclitus (Linnaeus, 1766) from in­ Mann, H., and E.M.V. Nambudiri. 2005. Charophytes of sular Newfoundland. Proceedings of the Helmintho­ lo­ ­ insular Newfoundland II: Chara evoluta and Chara ca­ gical Society of Washington 42: 111–116. nescens. Canadian Field-Naturalist 119: 26–37. https:// Endangered Species List Regulations. 2002. Newfound­ doi.org/10.22621/cfn.v119i1.77 land and Labrador Regulation 57/02. Accessed 11 Oc­to- Mitchell, J.S., and C.F. Purchase. 2014. Rapid coloniza- ­ber 2020. https://assembly.nl.ca/Legislation/sr/regulations/ tion of a species at risk: a new eastern range limit for rc020057.htm. Fundulus diaphanus (banded killifish), in found­New­ Fisheries and Oceans Canada. 2011. Management plan for land. Northeastern Naturalist 21: N41–N44. https://doi. the Banded Killifish Fundulus( diaphanus), New­found­ org/10.1656/045.021.0312 land population, in Canada. Species at Risk Act Man­age­ SARA (Species at Risk Act) Registry. 2019a. Species sum- ment Plan Series. Department of Fisheries and Oceans mary: Blackstripe Topminnow (Fundulus notatus). Ac­ Canada, Ottawa, Ontario, Canada. cessed 6 July 2020. https://species-registry.canada.ca/ Fritz, E.S., and E.T. Garside. 1974. Identification and de- index-en.html#/species/91-375. scription of hybrids of Fundulus heteroclitus and F. di­ SARA (Species at Risk Act) Registry. 2019b. Species sum- aphanus (Pisces: Cyprinodontidae) from Porter’s Lake, mary: Banded Killifish Fundulus ( diaphanus), New­ Nova Scotia, with evidence for absence of backcrossing. foundland populations). Accessed 6 July 2020. https:// Canadian Journal of Zoology 52: 1433–1442. https://doi. species-registry.canada.ca/index-en.html#/species/ org/10.1139/z74-184 85-321. Gibson, R.J., J.P. Thonney, and K. Hillier. 1984. An east- Scott, W.B., and E.J. Crossman. 1964. Fishes Occurring in erly extension in the known range for Fundulus diapha­ the Fresh Waters of Insular Newfoundland. Department 2020 Sargent et al.: Newfoundland killifish distributions 315

of Fisheries of Canada. Queen’s Printer and Controller of van Vliet, W.H. 1970. Shore and freshwater fish collections Stationary, Ottawa, Ontario, Canada. from Newfoundland. Publications in Zoology No. 3. Scott, W.B., and E.J. Crossman. 1973. Freshwater fishes National Museums of Canada, Ottawa, Ontario, Canada. of Canada. Bulletin 184. Fisheries Research Board of Willink, P.W., T.A. Widloe, V.J. Santucci, Jr., D. Makau­ Canada. Ottawa, Ontario, Canada. skas, J.S. Tiemann, S.D. Hertel, J.T. Lamer, and J.L. Scott, W.B., and M.G. Scott. 1988. Atlantic fishes of Can­ Sherwood. 2018. Rapid expansion of Banded Killi­fish ada. Canadian Bulletin of Fisheries and Aquatic Sciences Fundulus diaphanus across northern Illinois: dramatic re- No. 219. Ottawa, Ontario, Canada. covery or invasive species? American Midland­ Naturalist Systat. 2014. Sigmaplot for Windows version 13.0. Build 179: 179–190. https://doi.org/10.1674/0003-0031-179.2. 13.0.0.83. Systat Software, Inc. San Jose, California, USA. 179 Templeman, W. 1951. Report of the Newfoundland Fish­ eries Research Station for 1951. Fisheries Research Received 15 October 2019 Board of Canada, St. John’s, Newfoundland and Labra­ Accepted 21 October 2020 dor, Canada. Associate Editor: F. Chapleau

Supplementary Material: Table S1. List of museum collection records of Banded Killifish (Fundulus diaphanus), Mummichog (Fundulus heterocli­ tus), and potential hybrids (Fundulus sp.) from insular Newfoundland. Table S2. Meristics and morphometrics of Banded Killifish Fundulus( diaphanus), Mummichog (Fundulus heteroclitus), and potential hybrids (Fundulus sp.), collected throughout insular Newfoundland (2013–2016). Figure S1. Photographs used to verify presence of Banded Killifish Fundulus( diaphanus) and Mummichog (Fundulus heteroclitus) in insular Newfoundland where physical specimens could not be acquired. Figure S2. Frequency distributions comparing meristic and morphological characters of Banded Killifish Fundulus ( diaphanus), Mummichog (Fundulus heteroclitus), and potential hybrids (Fundulus sp.) collected in insular Newfoundland (2013–2016). Figure S3. Locations where Banded Killifish Fundulus( diaphanus) were not detected in insular Newfoundland. Figure S4. Locations where Mummichog (Fundulus heteroclitus) were not detected in insular Newfoundland.